temporalio 0.0.0 → 0.0.1

data/bridge/sdk-core/sdk/src/payload_converter.rs

@@ -0,0 +1,11 @@
+//! Very much subject to change. Defines traits for converting inputs/outputs to/from payloads
+
+use serde::{Deserialize, Serialize};
+
+/// Something that can be deserialized from a payload. Currently just a pass-through to
+/// [Deserialize] which may actually be the best long-run choice.
+pub trait FromPayload<'de>: Deserialize<'de> {}
+
+/// Something that can be serialized into a payload. Currently just a pass-through to
+/// [Serialize] which may actually be the best long-run choice.
+pub trait IntoPayload: Serialize {}

data/bridge/sdk-core/sdk/src/workflow_context/options.rs

@@ -0,0 +1,295 @@
+use std::{collections::HashMap, time::Duration};
+
+use temporal_client::WorkflowOptions;
+use temporal_sdk_core_protos::{
+    coresdk::{
+        child_workflow::ChildWorkflowCancellationType,
+        workflow_commands::{
+            ActivityCancellationType, ScheduleActivity, ScheduleLocalActivity,
+            StartChildWorkflowExecution,
+        },
+    },
+    temporal::api::common::v1::{Payload, RetryPolicy},
+};
+
+// TODO: Before release, probably best to avoid using proto types entirely here. They're awkward.
+
+pub trait IntoWorkflowCommand {
+    type WFCommandType;
+
+    /// Produces a workflow command from some options
+    fn into_command(self, seq: u32) -> Self::WFCommandType;
+}
+
+/// Options for scheduling an activity
+#[derive(Default, Debug)]
+pub struct ActivityOptions {
+    /// Identifier to use for tracking the activity in Workflow history.
+    /// The `activityId` can be accessed by the activity function.
+    /// Does not need to be unique.
+    ///
+    /// If `None` use the context's sequence number
+    pub activity_id: Option<String>,
+    /// Type of activity to schedule
+    pub activity_type: String,
+    /// Input to the activity
+    pub input: Payload,
+    /// Task queue to schedule the activity in
+    pub task_queue: String,
+    /// Time that the Activity Task can stay in the Task Queue before it is picked up by a Worker.
+    /// Do not specify this timeout unless using host specific Task Queues for Activity Tasks are
+    /// being used for routing.
+    /// `schedule_to_start_timeout` is always non-retryable.
+    /// Retrying after this timeout doesn't make sense as it would just put the Activity Task back
+    /// into the same Task Queue.
+    pub schedule_to_start_timeout: Option<Duration>,
+    /// Maximum time of a single Activity execution attempt.
+    /// Note that the Temporal Server doesn't detect Worker process failures directly.
+    /// It relies on this timeout to detect that an Activity that didn't complete on time.
+    /// So this timeout should be as short as the longest possible execution of the Activity body.
+    /// Potentially long running Activities must specify `heartbeat_timeout` and heartbeat from the
+    /// activity periodically for timely failure detection.
+    /// Either this option or `schedule_to_close_timeout` is required.
+    pub start_to_close_timeout: Option<Duration>,
+    /// Total time that a workflow is willing to wait for Activity to complete.
+    /// `schedule_to_close_timeout` limits the total time of an Activity's execution including
+    /// retries (use `start_to_close_timeout` to limit the time of a single attempt).
+    /// Either this option or `start_to_close_timeout` is required.
+    pub schedule_to_close_timeout: Option<Duration>,
+    /// Heartbeat interval. Activity must heartbeat before this interval passes after a last
+    /// heartbeat or activity start.
+    pub heartbeat_timeout: Option<Duration>,
+    /// Determines what the SDK does when the Activity is cancelled.
+    pub cancellation_type: ActivityCancellationType,
+    /// Activity retry policy
+    pub retry_policy: Option<RetryPolicy>,
+}
+
+impl IntoWorkflowCommand for ActivityOptions {
+    type WFCommandType = ScheduleActivity;
+    fn into_command(self, seq: u32) -> ScheduleActivity {
+        ScheduleActivity {
+            seq,
+            activity_id: match self.activity_id {
+                None => seq.to_string(),
+                Some(aid) => aid,
+            },
+            activity_type: self.activity_type,
+            task_queue: self.task_queue,
+            schedule_to_close_timeout: self
+                .schedule_to_close_timeout
+                .and_then(|d| d.try_into().ok()),
+            schedule_to_start_timeout: self
+                .schedule_to_start_timeout
+                .and_then(|d| d.try_into().ok()),
+            start_to_close_timeout: self.start_to_close_timeout.and_then(|d| d.try_into().ok()),
+            heartbeat_timeout: self.heartbeat_timeout.and_then(|d| d.try_into().ok()),
+            cancellation_type: self.cancellation_type as i32,
+            arguments: vec![self.input],
+            retry_policy: self.retry_policy,
+            ..Default::default()
+        }
+    }
+}
+
+/// Options for scheduling a local activity
+#[derive(Default, Debug, Clone)]
+pub struct LocalActivityOptions {
+    /// Identifier to use for tracking the activity in Workflow history.
+    /// The `activityId` can be accessed by the activity function.
+    /// Does not need to be unique.
+    ///
+    /// If `None` use the context's sequence number
+    pub activity_id: Option<String>,
+    /// Type of activity to schedule
+    pub activity_type: String,
+    /// Input to the activity
+    pub input: Payload,
+    /// Retry policy
+    pub retry_policy: RetryPolicy,
+    /// Override attempt number rather than using 1.
+    /// Ideally we would not expose this in a released Rust SDK, but it's needed for test.
+    pub attempt: Option<u32>,
+    /// Override schedule time when doing timer backoff.
+    /// Ideally we would not expose this in a released Rust SDK, but it's needed for test.
+    pub original_schedule_time: Option<prost_types::Timestamp>,
+    /// Retry backoffs over this amount will use a timer rather than a local retry
+    pub timer_backoff_threshold: Option<Duration>,
+    /// How the activity will cancel
+    pub cancel_type: ActivityCancellationType,
+    /// Indicates how long the caller is willing to wait for local activity completion. Limits how
+    /// long retries will be attempted. When not specified defaults to the workflow execution
+    /// timeout (which may be unset).
+    pub schedule_to_close_timeout: Option<Duration>,
+    /// Limits time the local activity can idle internally before being executed. That can happen if
+    /// the worker is currently at max concurrent local activity executions. This timeout is always
+    /// non retryable as all a retry would achieve is to put it back into the same queue. Defaults
+    /// to `schedule_to_close_timeout` if not specified and that is set. Must be <=
+    /// `schedule_to_close_timeout` when set, if not, it will be clamped down.
+    pub schedule_to_start_timeout: Option<Duration>,
+    /// Maximum time the local activity is allowed to execute after the task is dispatched. This
+    /// timeout is always retryable. Either or both of `schedule_to_close_timeout` and this must be
+    /// specified. If set, this must be <= `schedule_to_close_timeout`, if not, it will be clamped
+    /// down.
+    pub start_to_close_timeout: Option<Duration>,
+}
+
+impl IntoWorkflowCommand for LocalActivityOptions {
+    type WFCommandType = ScheduleLocalActivity;
+    fn into_command(mut self, seq: u32) -> ScheduleLocalActivity {
+        // Allow tests to avoid extra verbosity when they don't care about timeouts
+        // TODO: Builderize LA options
+        self.schedule_to_close_timeout
+            .get_or_insert(Duration::from_secs(100));
+
+        ScheduleLocalActivity {
+            seq,
+            attempt: self.attempt.unwrap_or(1),
+            original_schedule_time: self.original_schedule_time,
+            activity_id: match self.activity_id {
+                None => seq.to_string(),
+                Some(aid) => aid,
+            },
+            activity_type: self.activity_type,
+            arguments: vec![self.input],
+            retry_policy: Some(self.retry_policy),
+            local_retry_threshold: self.timer_backoff_threshold.and_then(|d| d.try_into().ok()),
+            cancellation_type: self.cancel_type.into(),
+            schedule_to_close_timeout: self
+                .schedule_to_close_timeout
+                .and_then(|d| d.try_into().ok()),
+            schedule_to_start_timeout: self
+                .schedule_to_start_timeout
+                .and_then(|d| d.try_into().ok()),
+            start_to_close_timeout: self.start_to_close_timeout.and_then(|d| d.try_into().ok()),
+            ..Default::default()
+        }
+    }
+}
+
+/// Options for scheduling a child workflow
+#[derive(Default, Debug, Clone)]
+pub struct ChildWorkflowOptions {
+    /// Workflow ID
+    pub workflow_id: String,
+    /// Type of workflow to schedule
+    pub workflow_type: String,
+    /// Input to send the child Workflow
+    pub input: Vec<Payload>,
+    /// Cancellation strategy for the child workflow
+    pub cancel_type: ChildWorkflowCancellationType,
+    /// Common options
+    pub options: WorkflowOptions,
+}
+
+impl IntoWorkflowCommand for ChildWorkflowOptions {
+    type WFCommandType = StartChildWorkflowExecution;
+    fn into_command(self, seq: u32) -> StartChildWorkflowExecution {
+        StartChildWorkflowExecution {
+            seq,
+            workflow_id: self.workflow_id,
+            workflow_type: self.workflow_type,
+            input: self.input,
+            cancellation_type: self.cancel_type as i32,
+            workflow_id_reuse_policy: self.options.id_reuse_policy as i32,
+            workflow_execution_timeout: self
+                .options
+                .execution_timeout
+                .and_then(|d| d.try_into().ok()),
+            workflow_run_timeout: self
+                .options
+                .execution_timeout
+                .and_then(|d| d.try_into().ok()),
+            workflow_task_timeout: self.options.task_timeout.and_then(|d| d.try_into().ok()),
+            search_attributes: self.options.search_attributes.unwrap_or_default(),
+            cron_schedule: self.options.cron_schedule.unwrap_or_default(),
+            ..Default::default()
+        }
+    }
+}
+
+/// Options for sending a signal to an external workflow
+pub struct SignalWorkflowOptions {
+    /// The workflow's id
+    pub workflow_id: String,
+    /// The particular run to target, or latest if `None`
+    pub run_id: Option<String>,
+    /// The details of the signal to send
+    pub signal: Signal,
+}
+
+impl SignalWorkflowOptions {
+    /// Create options for sending a signal to another workflow
+    pub fn new(
+        workflow_id: impl Into<String>,
+        run_id: impl Into<String>,
+        name: impl Into<String>,
+        input: impl IntoIterator<Item = impl Into<Payload>>,
+    ) -> Self {
+        Self {
+            workflow_id: workflow_id.into(),
+            run_id: Some(run_id.into()),
+            signal: Signal::new(name, input),
+        }
+    }
+
+    /// Set a header k/v pair attached to the signal
+    pub fn with_header(
+        &mut self,
+        key: impl Into<String>,
+        payload: impl Into<Payload>,
+    ) -> &mut Self {
+        self.signal.data.with_header(key.into(), payload.into());
+        self
+    }
+}
+
+/// Information needed to send a specific signal
+pub struct Signal {
+    /// The signal name
+    pub signal_name: String,
+    /// The data the signal carries
+    pub data: SignalData,
+}
+
+impl Signal {
+    /// Create a new signal
+    pub fn new(
+        name: impl Into<String>,
+        input: impl IntoIterator<Item = impl Into<Payload>>,
+    ) -> Self {
+        Self {
+            signal_name: name.into(),
+            data: SignalData::new(input),
+        }
+    }
+}
+
+/// Data contained within a signal
+#[derive(Default)]
+pub struct SignalData {
+    /// The arguments the signal will receive
+    pub input: Vec<Payload>,
+    /// Metadata attached to the signal
+    pub headers: HashMap<String, Payload>,
+}
+
+impl SignalData {
+    /// Create data for a signal
+    pub fn new(input: impl IntoIterator<Item = impl Into<Payload>>) -> Self {
+        Self {
+            input: input.into_iter().map(Into::into).collect(),
+            headers: HashMap::new(),
+        }
+    }
+
+    /// Set a header k/v pair attached to the signal
+    pub fn with_header(
+        &mut self,
+        key: impl Into<String>,
+        payload: impl Into<Payload>,
+    ) -> &mut Self {
+        self.headers.insert(key.into(), payload.into());
+        self
+    }
+}